The development of human cancer is complex process involving the acquisition of multiple genetic changes that confer a selective advantage on a tumor clone. Many of the genes involved in these genetic changes have been identified based on mutation in cancer in model organisms, inheritance in a mutant form in cancer- prone families, or presence at recurrent translocation breakpoints. However, a very large number of chromosome abnormalities have been detected for which no corresponding molecular explanation has been provided at the level of individual genes. This includes a very large class of unbalanced chromosomal abnormalities in which there is a net loss or gain of genetic material. Such unbalanced chromosomal abnormalities are especially common in solid tumors. A molecular explanation of these changes is highly desirable for breakthroughs in our understanding of the cancer process and for identification of novel targets for therapeutic exploration. Many strategies for understanding the complex genetic picture presented by the recurrent unbalanced chromosome abnormalities in human cancer have been proposed; including array- based comparative genome hybridization, and microarray-based expression studies. We plan to use a functional approach. We have utilized the Sleeping Beauty (SB) transposable element for forward genetic screens for cancer in mice. This project will be extended upon and be used in new unbiased screens to define genetic changes that can cause cancer and create useful in vivo models for future studies. [unreadable] [unreadable] [unreadable]